Derived categories of quartic double fivefolds

This paper constructs specific singular and rational quartic double fivefolds whose Kuznetsov components admit crepant categorical resolutions by (twisted or untwisted) Calabi–Yau threefolds, thereby providing evidence for higher-dimensional rationality conjectures and the connectedness of Calabi–Yau moduli spaces.

The Gist

Imagine you are an architect trying to understand the "blueprint" of a very strange, high-dimensional building. In the world of mathematics, these buildings are called varieties, and the blueprints are called derived categories. These blueprints tell you everything about the building's shape, its holes, and how its parts fit together.

This paper is about a specific, tricky building called a Quartic Double Fivefold.

Here is the story of what the authors did, explained without the heavy math jargon.

1. The Problem: A Building with a Cracked Foundation

Imagine a building that is supposed to be perfectly smooth and symmetrical (a "smooth" variety). Mathematicians have a theory (Kuznetsov's Conjecture) that says: If a building is "rational" (meaning it can be built from simple blocks like a cube), then its complex blueprint must secretly look like the blueprint of a simpler, 3D garden (a Calabi-Yau threefold).

However, the authors looked at their specific building (the Quartic Double Fivefold) and realized something scary: It's impossible for this building to be smooth and rational at the same time. If they tried to force the blueprint to look like a simple garden, the math broke. It was like trying to fit a square peg in a round hole.

Furthermore, no one knew if this specific building was "rational" or not. It was a mystery.

2. The Strategy: Intentionally Breaking the Building

Instead of trying to fix the building to make it smooth, the authors decided to intentionally break it.

They looked at versions of the building that had a specific, controlled crack (a singularity) running through it. Think of it like taking a perfect sphere and crushing it slightly along a line.

• The Twist: When you crush the building in this specific way, the "blueprint" (the derived category) changes. It becomes messy, but it becomes solvable.
• The Result: They found that the messy blueprint of this broken building could be "cleaned up" (resolved) into a new, perfect blueprint.

3. The Discovery: Two Types of Cleanups

The authors found two different ways to clean up the blueprint, depending on how they broke the building.

Case A: The "Twisted" Garden (The General Case)

In the first scenario, they broke the building along a line. To fix the blueprint, they had to build a new 3D garden.

• The Catch: This garden wasn't a normal garden. It was a "Twisted" garden. Imagine a garden where the flowers are arranged in a pattern that requires a special key (a "Brauer class") to unlock. You can see the garden, but you can't walk through it freely without that key.
• The Metaphor: It's like a 3D world that exists inside a video game with a glitch. The world is beautiful and follows the rules of a "Calabi-Yau" (a special type of geometry), but it's "twisted" so you can't interact with it directly.

Case B: The "Perfect" Garden (The Special Case)

In the second scenario, they broke the building even more specifically. They made sure the crack aligned with a hidden "secret passage" (a section of the building).

• The Result: Because of this secret passage, the "Twisted" garden from Case A suddenly lost its twist. The key disappeared!
• The Metaphor: The garden is now a normal, physical 3D garden. You can walk through it, touch the flowers, and it's perfectly smooth.
• The Big Win: Because this garden is now a normal, physical object, the authors could prove that the original broken building was indeed rational. They found the "simple blocks" that built it.

4. Why This Matters: The "Fantasy" of Connected Worlds

The paper connects to a famous idea in math called "Reid's Fantasy."

• The Fantasy: Imagine that all possible 3D Calabi-Yau gardens are connected by a network of tunnels. You can start in one garden, walk through a tunnel (a "conifold transition"), and end up in a completely different garden. This suggests that all these complex shapes are just different views of the same underlying reality.
• The Paper's Contribution: The authors showed that you can do this not just with physical gardens, but with these abstract, "non-commutative" blueprints too.
  1. They started with a smooth building.
  2. They broke it to get a "Twisted" garden (Case A).
  3. They tweaked the break to get a "Perfect" garden (Case B).
  4. This proves that even in the weird, abstract world of math, these different shapes are connected.

Summary Analogy

Think of the Quartic Double Fivefold as a complex, 5D sculpture.

• The Goal: To see if this sculpture is made of simple Lego bricks (Rationality).
• The Obstacle: The sculpture is too complex to see the bricks directly.
• The Solution: The authors smashed the sculpture in a specific way.
  • First smash: The pieces reassemble into a hologram (the Twisted Garden). It looks real, but it's not quite solid.
  • Second smash: The hologram solidifies into a real statue (the Geometric Garden).
• The Conclusion: Because the pieces could solidify into a real statue, the original sculpture must have been made of Lego bricks all along.

This paper is a triumph because it uses "controlled destruction" to prove that a complex mathematical object is actually simple, and it shows how different mathematical worlds are connected by these transformations.

Technical Summary

Here is a detailed technical summary of the paper "Derived Categories of Quartic Double Fivefolds" by Raymond Cheng, Alexander Perry, and Xiaolei Zhao.

1. Problem Statement and Context

The paper addresses the Kuznetsov rationality conjecture and Reid's fantasy in the context of higher-dimensional algebraic geometry.

• The Kuznetsov Component: For a smooth prime Fano variety $X$ of index $r$, the bounded derived category of coherent sheaves $D^b(X)$ admits a semiorthogonal decomposition:
  $$D^b(X) = \langle \text{Ku}(X), \mathcal{O}_X, \mathcal{O}_X(1), \dots, \mathcal{O}_X(r-1) \rangle$$
  The subcategory $\text{Ku}(X)$ is called the Kuznetsov component. It is expected to capture the "interesting" geometric part of $X$.
• The Conjecture: Kuznetsov conjectured that if $X$ is a rational Fano variety, then $\text{Ku}(X)$ is equivalent to the derived category of a smooth projective Calabi–Yau variety of dimension $\dim(X)-2$.
• The Specific Case: The authors focus on quartic double fivefolds ($X \to \mathbb{P}^5$ branched along a quartic hypersurface). Here, $\dim(X)=5$ and the index is $r=4$, so $\text{Ku}(X)$ is expected to be a Calabi–Yau category of dimension 3.
• The Obstacle: It is known (via Hochschild homology computations) that for a smooth quartic double fivefold, $\text{Ku}(X)$ is not equivalent to the derived category of any smooth projective variety. This implies that if the conjecture holds, smooth quartic double fivefolds must be irrational (a major open problem).
• The Goal: Instead of attacking the smooth case directly, the authors investigate singular quartic double fivefolds. They aim to construct specific singular examples where $\text{Ku}(X)$ admits a crepant categorical resolution by a geometric Calabi–Yau threefold (either twisted or untwisted), thereby providing evidence for a higher-dimensional, noncommutative version of Reid's fantasy (the connectedness of the moduli space of Calabi–Yau threefolds).

2. Methodology

The authors employ a combination of birational geometry, derived category techniques (semiorthogonal decompositions, mutations), and Clifford algebra theory.

1. Geometric Setup:

   • They consider a double cover $X \to \mathbb{P}^5$ branched along a quartic hypersurface $Y \subset \mathbb{P}^5$ that is singular along a line $L$.
   • They resolve the singularities of $X$ by blowing up the line $L$ in $\mathbb{P}^5$ to get $\tilde{\mathbb{P}}^5$, and then taking the strict transform $\tilde{X}$.
   • The resolution $\tilde{X}$ is realized as a quadric surface bundle over $\mathbb{P}^5$ (specifically, a double cover of a $\mathbb{P}^2$-bundle).

2. Categorical Resolution:

   • They define a "Kuznetsov component" $\widetilde{\text{Ku}}(X) \subset D^b(\tilde{X})$ for the singular variety.
   • Using theorems by Kuznetsov regarding resolutions of singularities, they establish that $\widetilde{\text{Ku}}(X)$ serves as a crepant categorical resolution of $\text{Ku}(X)$.

3. Clifford Algebra and Quadric Bundles:

   • The derived category of the quadric bundle $\tilde{X}$ is described using the sheaf of even Clifford algebras $B_0$ associated with the bundle.
   • They relate $\widetilde{\text{Ku}}(X)$ to the derived category of the base $\mathbb{P}^5$ with coefficients in $B_0$, denoted $D^b(\mathbb{P}^5, B_0)$.

4. Mutation Techniques:

   • A significant portion of the technical work involves performing a sequence of mutations on semiorthogonal decompositions. They transform the decomposition of $D^b(\tilde{X})$ arising from the quadric bundle structure into one that isolates the Kuznetsov component, allowing them to identify it with specific geometric objects.

5. Specialization to Rationality:

   • They specialize the geometry to a case where the branch locus $Y$ is tangent to a complementary 3-plane $P$ along a smooth quadric surface $S$. This geometric condition ensures the existence of a rational section for the quadric bundle, which is crucial for removing the "twist" (Brauer class) in the categorical resolution.

3. Key Contributions and Results

The paper presents two main theorems corresponding to two levels of specialization:

Theorem 1.3: The Twisted Case (General Singular)

• Setup: $X$ is a double cover branched along a quartic $Y$ singular along a line $L$.
• Result: There exists a crepant categorical resolution of $\text{Ku}(X)$ by $D^b(W^+, \mathcal{A}^+)$.
  • $W^+$ is a proper 3-dimensional Calabi–Yau algebraic space (not projective).
  • $\mathcal{A}^+$ is an Azumaya algebra on $W^+$ with a non-trivial Brauer class.
• Significance: This confirms that the Kuznetsov component of a singular quartic double fivefold is equivalent to a "twisted" Calabi–Yau threefold. The non-triviality of the Brauer class is linked to the non-existence of a rational section for the associated quadric bundle.
• Non-projectivity: The authors prove $W^+$ is not projective by showing its "defect" (related to the difference between Betti numbers $b_4$ and $b_2$) vanishes, implying no ample divisors exist.

Theorem 1.5: The Untwisted Rational Case (Specialized)

• Setup: $X$ is as above, but $Y$ is further specialized to be tangent to a 3-plane $P$ along a smooth quadric $S$.
• Result:
  1. There exists a crepant categorical resolution of $\text{Ku}(X)$ by $D^b(W^{++})$, where $W^{++}$ is a projective Calabi–Yau threefold (no twist/Azumaya algebra).
  2. In this specific configuration, the variety $X$ is rational.
• Mechanism: The tangency condition provides a rational section for the quadric bundle. This section allows the authors to perform a "hyperbolic reduction" (a geometric operation reducing the fiber dimension) and construct a small resolution $W^{++}$ that is smooth and projective.
• Significance: This provides a concrete instance where a rational Fano variety has a Kuznetsov component equivalent to the derived category of a smooth projective Calabi–Yau threefold, validating the higher-dimensional Kuznetsov conjecture for this specific singular case.

4. Significance and Implications

1. Higher-Dimensional Kuznetsov Conjecture: The paper provides the first explicit examples in dimension 5 supporting the conjecture that rationality of a Fano variety implies its Kuznetsov component is equivalent to a Calabi–Yau category.
2. Noncommutative Reid's Fantasy: The results support the "noncommutative" version of Reid's fantasy (proposed by Kuznetsov and Perry). They demonstrate a path connecting the Kuznetsov component of a smooth (but potentially irrational) fivefold to a twisted Calabi–Yau, and then to a geometric Calabi–Yau via degeneration and crepant resolution. This suggests that all such categories (and their moduli spaces) are connected in a "fantasy" of transitions.
3. Rationality of Singular Varieties: The construction of rational specializations of quartic double fivefolds is a significant step in the rationality problem, a notoriously difficult area in algebraic geometry.
4. Methodological Advancement: The paper refines the technique of using crepant categorical resolutions for singular varieties. It demonstrates how to handle the transition from twisted (noncommutative) to untwisted (commutative) categories via geometric degenerations, offering a blueprint for studying derived categories of other high-dimensional Fano varieties.

In summary, the paper successfully constructs a bridge between the derived category of a singular Fano fivefold and the geometry of Calabi–Yau threefolds, confirming deep conjectures about the structure of rational varieties and the connectivity of Calabi–Yau moduli spaces in a noncommutative setting.